This article was downloaded by: [Ryerson University] On: 03 February 2015, At: 00:51 Publisher: Taylor & Francis Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK
Food Additives & Contaminants: Part A Publication details, including instructions for authors and subscription information: http://www.tandfonline.com/loi/tfac20
A survey of bisphenol A and other bisphenol analogues in foodstuffs from nine cities in China a
Chunyang Liao & Kurunthachalam Kannan
a
a
Wadsworth Center, New York State Department of Health, and Department of Environmental Health Sciences, School of Public Health, State University of New York at Albany, Empire State Plaza, Albany, NY 12201-0509, USA Accepted author version posted online: 21 Nov 2013.Published online: 30 Jan 2014.
Click for updates To cite this article: Chunyang Liao & Kurunthachalam Kannan (2014) A survey of bisphenol A and other bisphenol analogues in foodstuffs from nine cities in China, Food Additives & Contaminants: Part A, 31:2, 319-329, DOI: 10.1080/19440049.2013.868611 To link to this article: http://dx.doi.org/10.1080/19440049.2013.868611
PLEASE SCROLL DOWN FOR ARTICLE Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) contained in the publications on our platform. However, Taylor & Francis, our agents, and our licensors make no representations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of the Content. Any opinions and views expressed in this publication are the opinions and views of the authors, and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon and should be independently verified with primary sources of information. Taylor and Francis shall not be liable for any losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoever or howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use of the Content. This article may be used for research, teaching, and private study purposes. Any substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any form to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http:// www.tandfonline.com/page/terms-and-conditions
Food Additives & Contaminants: Part A, 2014 Vol. 31, No. 2, 319–329, http://dx.doi.org/10.1080/19440049.2013.868611
A survey of bisphenol A and other bisphenol analogues in foodstuffs from nine cities in China Chunyang Liao and Kurunthachalam Kannan* Wadsworth Center, New York State Department of Health, and Department of Environmental Health Sciences, School of Public Health, State University of New York at Albany, Empire State Plaza, Albany, NY 12201-0509, USA
Downloaded by [Ryerson University] at 00:51 03 February 2015
(Received 29 October 2013; accepted 19 November 2013) Bisphenol A (BPA) is a high-production-volume chemical that is widely used in polycarbonate plastics and epoxy food-can coatings. Following several studies that have reported adverse effects of BPA over the past decade, other bisphenol analogues, such as bisphenol F (BPF), bisphenol S (BPS), bisphenol AF (BPAF), and bisphenol B (BPB), have been gradually developed as substitutes for BPA in several applications. Nevertheless, few studies have reported on the occurrence of compounds other than BPA in foodstuffs. In this study, 289 food samples (13 categories: cereals and cereal products, meat and meat products, fish and seafood, eggs, milk and milk products, bean products, fruits, vegetables, cookies/snacks, beverages, cooking oils, condiments, and others), collected from nine cities in China, were analysed for eight bisphenol analogues using high-performance liquid chromatography-tandem mass spectrometry (HPLC-MS/MS). BPA and BPF were found widely in foodstuffs at concentrations ranging from below the limit of quantitation (LOQ) to 299 ng g–1 (mean = 4.94 ng g–1) and from below the LOQ to 623 ng g–1 (mean = 2.50 ng g–1), fresh weight, respectively. The highest total concentrations of bisphenols (∑BPs: sum of eight bisphenols) were found in the category of vegetables that included canned products (mean = 27.0 ng g–1), followed by fish and seafood (16.5 ng g–1) and beverages (15.6 ng g–1). ∑BP concentrations (mean = 2–3 ng g–1) in milk and milk products, cooking oils, and eggs were low. Food samples sold in metallic cans contained higher mean ∑BP concentrations (56.9 ng g–1) in comparison with those packaged in glass (0.43 ng g–1), paper (11.9 ng g–1), or plastic (6.40 ng g–1). The daily dietary intakes of bisphenols were estimated, based on the mean concentrations measured and daily consumption rates of foods, to be 646 and 664 ng kg–1 bw day–1 for men and women, respectively. Keywords: bisphenols; BPA; food; dietary exposure; market basket survey; daily intake
Introduction Bisphenol A (BPA) is widely used as a monomer in the manufacturing of polycarbonate plastics and epoxy resins, which are used in a wide range of consumer products, including plastic food containers and epoxy food-can coatings (Geens et al. 2012; Huang et al. 2012; Vandenberg et al. 2012). BPA is also used in printed circuit boards, building materials, compact discs, medical devices, dental fillings and thermal receipt papers (Geens et al. 2012; Gallart-Ayala et al. 2013). It has been estimated that approximately 3% of the polycarbonate and 10% of the epoxy resins are used in materials intended to come in contact with foods (Plastics Europe 2007). Other bisphenol analogues, including bisphenol F (BPF), bisphenol S (BPS), bisphenol AF (BPAF), and bisphenol B (BPB), have been gradually developed as substitutes for BPA in the production of polycarbonate plastics and epoxy resins (Sueiro et al. 2003; Satoh et al. 2004; Kuruto-Niwa et al. 2005; Feng et al. 2012). In comparison with BPA, BPF possesses lower viscosity and better resistance against solvents and is increasingly used in epoxy resin products (Danzl et al. 2009). BPS is widely used as a replacement for BPA in thermal receipt papers (Liao Liu, Kannan, et al. 2012). BPAF is used as a cross-linking reagent in the *Corresponding author. Email: [email protected] © 2014 Taylor & Francis
production of fluoropolymers and fluoroelastomers (Feng et al. 2012). Considerable controversy surrounds exposure to and toxicity of BPA, and this compound has been identified as an endocrine disruptor due to its potential to elicit developmental and reproductive toxicity (Geens et al. 2012; Huang et al. 2012; Vandenberg et al. 2012). Adverse effects of BPA at environmentally relevant exposure doses have been reported in laboratory animals (Vandenberg et al. 2012). Neonatal exposure to BPA of rats at 10 µg kg–1 body weight (bw) resulted in early and persistent over-expression of genes involved in DNA methylation/demethylation (Tang et al. 2012). In an in vitro study using murine cells, exposure to BPA at concentrations ranging from 10−10 to 10−8 M (approximately 0.023 to 2.3 ng ml–1) markedly induced spermatogonial proliferation (Sheng & Zhu 2011). Acute toxicity, genotoxicity and estrogenic activity of BPS, BPB and BPF have been reported (Chen et al. 2002; Rivas et al. 2002; Okuda et al. 2011; Pisapia et al. 2012). One study showed that exposure to BPAF considerably reduced testosterone levels in adult male rats (Feng et al. 2012). A few other studies have shown that BPF and BPS are more resistant to degradation in the environment than is BPA (Ike et al. 2006; Danzl et al. 2009).
Downloaded by [Ryerson University] at 00:51 03 February 2015
320
C. Liao and K. Kannan
Humans can be exposed to BPA that leaches from polycarbonate plastics or epoxy resin packaging materials into foodstuffs (Vandenberg et al. 2007). Earlier studies have shown that BPA can leach from polycarbonate drinking and baby bottles or from metallic-can epoxy coatings into various food products (Cao & Corriveau 2008; Cao et al. 2009; Cooper et al. 2011; Simoneau et al. 2011). Heat processing or acidic conditions can enhance the leaching of BPA from packaging materials or storage containers (Munguia-Lopez & Soto-Valdez 2001; Munguia-Lopez et al. 2002; Lim et al. 2009). In view of the migration of BPA from packaging materials, the European Commission has established a migration limit of 0.6 mg kg–1 for BPA in plastic materials and articles intended to come in contact with foodstuffs (European Commission 2004). Although very few studies have reported on the migration of bisphenols other than BPA from plastics and epoxy coatings (Viñas et al. 2010; Oca et al. 2013), it is expected to be similar to those reported for BPA. A few pilot studies have reported the occurrence of BPA in food items collected in China (Shao, Han, Li, et al. 2007; Shao, Han, Tu, et al. 2007; Niu et al. 2012). BPA was found in 13 of 27 (48%) meat samples, including pork, beef, lamb, chicken, duck and fish, collected from supermarkets in Beijing, China, at concentrations ranging from 0.33 to 7.08 ng g–1 wet weight (ww) (Shao, Han, Li, et al. 2007). Low concentrations and detection rates for BPA were reported in milk (10%) and eggs (30%) from China, at concentrations ranging from 0.35 to 10.5 ng g–1 ww (Shao, Han, Tu, et al. 2007). A recent study reported BPA in several cereals and cereal products from Beijing and found BPA at measurable levels (range = 1.0– 3.8 ng g–1 ww; detection rate = 21%) (Niu et al. 2012). The earlier studies were sporadic and analysed very few samples collected from localised areas in China. To our knowledge, studies on the occurrence of other bisphenol analogues in foodstuffs from China are not available. The aim of this study was to assess comprehensively dietary exposure to bisphenols by the general population in China. A total of 289 food samples were collected from nine cities in China in 2012 for the analysis of eight bisphenol analogues, including BPA, BPAF, BPB, BPF and BPS. Concentrations of bisphenols in several food categories and in foodstuffs sold in various types of packaging materials were determined and regional distribution in concentrations elucidated. On the basis of the measured concentrations in food samples and reported daily consumption rates of individual food items, the dietary intakes of bisphenols were calculated.
(BPAP; 1571-75-1; 99%), bisphenol F (BPF; 620-92-8; 98%), bisphenol P (BPP; 2167-51-3; 99%), bisphenol S (BPS; 80-09-1; 98%), and bisphenol Z (BPZ; 843-55-0; 98%) were purchased from Sigma-Aldrich (St. Louis, MO, USA). Bisphenol B (BPB; 77-40-7; 98%) was purchased from TCI America (Portland, OR, USA); and 13 C12-labelled BPA (263261-65-0; 99%) was from Cambridge Isotope Laboratories (Andover, MA, USA). Formic acid (98.2%) was purchased from Sigma-Aldrich. All organic solvents (HPLC grade), including methanol, acetonitrile, acetone, hexane and dichloromethane, were purchased from Mallinckrodt Baker (Phillipsburg, NJ, USA). Milli-Q water was prepared by an ultrapure water system (Barnstead International, Dubuque, IA, USA). Sample collection From July to September 2012, a total of 289 food samples were collected from nine cities in China, including Baoding (n = 28), Beijing (31), Harbin (41), Jinan (41), Jinchang (36), Liuzhou (49), Qinghuangdao (37), Shanghai (14) and Tianjin (12) (Figure S1 in the Supplementary material). Food samples, purchased mainly from large retail stores and a few local grocery stores, were grouped into 13 categories as cereals and cereal products (rice and rice products/wheat flour/bread/noodles/pasta/corn products), meat and meat products (beef/ pork/chicken/duck/sausages), fish and seafood (fish/ shrimp/squid), eggs (salted eggs), milk and milk products (milk/infant formula/yogurt/cheese), bean products (red and green beans/soybeans/orchid beans/dried bean curd), fruits (walnuts/chestnuts/jujubes/plums/hawthorn/raisins), vegetables (mushrooms/peanuts/peppers/seaweed, bamboo shoots/potatoes/cabbage/salted mustard), cookies/ snacks (candy/chocolate/biscuits/potato chips/pancakes), beverages (juice/liquor/coffee drinks), cooking oils (bean, peanut, corn, and sesame oils), condiments (soy sauce/ vinegar/cooking wine/ketchup/bean paste/aniseed/chilli powder), and others (jelly/black sesame powder/lotus root starch/milk tea powder/coffee powder). Various brands of foods, including national brands, store brands and specialty brands, were chosen to represent items commonly consumed by the Chinese population. Further details of the food items analysed in this study have been reported elsewhere (Liao et al. 2013). All samples were stored at –20°C until processing. Sample preparation
Materials and methods Chemicals Bisphenol A (BPA; CAS No.: 80-05-7; purity: 97%), bisphenol AF (BPAF; 1478-61-1; 97%), bisphenol AP
The extraction and clean-up procedures in the analysis of food samples have been described in detail elsewhere (Liao & Kannan 2013). For the purpose of extraction, food samples were divided into four broad categories: solid foods, cooking oils, beverages, and milk and its products. Solid foods were homogenised (using a stainless
Downloaded by [Ryerson University] at 00:51 03 February 2015
Food Additives & Contaminants: Part A steel Sorvall® Omni-Mixer; Dupont, Newtown, CT, USA), freeze dried, weighed (1.0–3.0 g dry weight), and extracted twice with acetonitrile (6 ml each). After evaporation to near dryness, the extract was redissolved in 2 ml of 10% dichloromethane/hexane and purified with a Strata® NH2 cartridge (200 mg/3 cc; Phenomenex, Torrance, CA, USA). The final elute was concentrated to 0.5 ml. The cooking oil samples (1.5–4.0 g) were extracted using the same procedure described above, without the freeze-drying step. Beverage samples were weighed (approximately 5 g) and extracted twice with ethyl acetate (6 ml each) by liquid–liquid extraction (LLE). The extract was evaporated to near dryness, redissolved in 2 ml of 10% dichloromethane/hexane (v/v), and purified with Strata® NH2 cartridge, as described above. Milk and milk products (approximately 3 g) were extracted with about 6 ml of acetonitrile (sample: acetonitrile = 1:2, v/v) (Yan et al. 2009) by LLE. After centrifugation, the supernatant was evaporated to approximately 4 ml, diluted to 10 ml with 0.2% formic acid (pH 2.5), and purified with an Oasis® MCX cartridge (60 mg/3 cc; Waters, Milford, MA, USA) (Liao & Kannan 2013). The final elute was concentrated to 0.5 ml. HPLC-MS/MS analysis A Shimadzu Prominence Series liquid chromatograph, LC-20AD system (Shimadzu USA, Canby, OR, USA) was used for the separation of target analytes. The analytical column was Betasil C18 (2.1 × 100 mm, 5 μm; Thermo Electron Corporation, Waltham, MA, USA), which was connected to a Javelin guard column (Betasil C18, 2.1 × 20 mm, 5 μm; Thermo Electron Corporation). The mobile phase was methanol and water; the elution gradient is shown in Table S1 in the Supplementary material. The mobile phase flow rate was 300 µl min–1, and 10 µl of the sample extract was injected. An Applied Biosystems API 3200 electrospray triple quadrupole mass spectrometer (ESI-MS/MS; Applied Biosystems, Foster City, CA, USA) was used for the determination of bisphenols. The MS/MS was operated in an electrospray-negative ionisation multiple reaction monitoring (MRM) mode, and the MRM transitions were optimised for each analyte (Table S2 in the Supplementary material). The MS/MS parameters are shown in Table S3 of the Supplementary material. Quality assurance and quality control (QA/QC) Identification of target analytes was based on the retention times relative to 13C12-BPA. Quantification was based on the linear regression curve (r > 0.99) generated from a 12point calibration standard at concentrations in the range of 0.01–100 ng ml–1. The measured concentrations were corrected for the recoveries of the internal standard. For
321
every 60 food samples analysed, procedural blanks (n = 3), spiked blanks (n = 3), and spiked matrixes (n = 4) were analysed to evaluate possible contamination and matrix effects from sample preparation steps and instrumental analysis. Procedural blanks contained trace levels of BPA (0.048 ng g–1), BPF (0.067 ng g–1), and BPS (0.005 ng g–1), and these values were subtracted (from mean values found in procedural blanks for each batch of analysis) from sample values. The recoveries of standards spiked in blanks (20 ng for each analyte; mean ± SD) ranged from 62% ± 7% (BPS) to 107% ± 18% (BPB). The average recoveries of standards spiked in food matrixes (20 ng for each analyte), from four broad categories of foods, were in the range of 79–109% for solid foods, 62–120% for cooking oils, 78–122% for beverages, and 66–126% for milk and milk products (Table S4 in the Supplementary material). The limit of quantitations (LOQs), calculated from the value of the lowest acceptable calibration standard and a nominal sample weight of 1.0 g, were 0.01 ng g–1 ww for BPA, BPAF, BPAP and BPS, 0.025 ng g–1 ww for BPB and BPP, and 0.05 ng g–1 ww for BPF and BPZ (Table S2 in the Supplementary material). A calibration standard and methanol were injected after every 10 samples on the instrument to check for instrumental background, carryover and stability.
Data analysis Liquid samples were weighed prior to extraction. The moisture contents of solid food samples were determined for the conversion of data from a dry-weight to a wetweight basis. All data are presented on a wet-weight (or fresh-weight) basis, unless specified otherwise. For statistical analysis, values below the LOQ were set to half the LOQ (LOQ/2). Spearman’s correlation analysis and oneway analysis of variance (ANOVA) with the Tukey test were conducted using SPSS 17.0 and Origin 7.5. A value of p < 0.05 was considered as being statistically significant.
Results and discussion Bisphenol concentrations BPA was frequently (detection rate = 60.9%) found in food samples from China at concentrations ranging from < LOQ to 299 ng g–1 fresh weight (mean = 4.94 ng g–1) (Table 1). The highest BPA concentrations (mean = 15.5, 95th percentile = 41.3 ng g–1) were found in beverages; this was followed, in decreasing order, by fish and seafood (14.1, 40.2 ng g–1), fruits (7.76, 14.5 ng g–1), and condiments (7.19, 12.6 ng g–1). Canned foods contributed a relatively high mean concentration of BPA in these food categories; in particular, three of four samples in the beverages
Cereals and cereal products (n = 39) Mean 5.60 Median 0.383 95th percentile 18.6 Range n.d.–130 Frequency (%) 69.2 Meat and meat products (n = 20) Mean 0.579 Median 0.098 95th percentile 2.26 Range n.d.–3.14 Frequency (%) 55.0 Fish and seafood (n = 11) Mean 14.1 Median 4.46 95th percentile 40.2 Range 0.327–42.1 Frequency (%) 100 Eggs (n = 11) Mean 2.15 Median 0.967 95th percentile 7.53 Range n.d.–9.22 Frequency (%) 63.6 Milk and milk products (n = 17) Mean 1.47 Median 0.942 95th percentile 3.76 Range n.d.–10.8 Frequency (%) 82.4 Bean products (n = 27) Mean 6.48 Median 0.147 95th percentile 46.1 Range n.d.–66.6 Frequency (%) 63.0 Fruits (n = 20) Mean 7.76 Median 0.530 95th percentile 14.5 Range n.d.–132 Frequency (%) 85.0 Vegetables (n = 42) Mean 2.88
BPA 0.027 0.005 0.033 n.d.–0.575 5.1 0.102 0.005 0.343 n.d.–1.67 10.0 0.006 0.005 0.011 n.d.–0.017 9.1 0.010 0.005 0.030 n.d.–0.047 18.2 0.063 0.005 0.352 n.d.–0.746 11.8 0.029 0.005 0.035 n.d.–0.622 7.4 0.007 0.005 0.019 n.d.–0.031 10.0 4.68
0.012 0.005 0.054 n.d.–0.097 10.0 0.088 0.005 0.446 n.d.–0.759 36.4 0.020 0.005 0.075 n.d.–0.117 36.4 0.005 0.005 0.005 n.d. 0 0.006 0.005 0.009 n.d.–0.014 7.4 0.008 0.005 0.018 n.d.–0.023 25.0 0.031
BPAP
0.005 0.005 0.005 n.d.–0.016 2.6
BPAF
0.013
0.013 0.013 0.013 n.d. 0
0.013 0.013 0.013 n.d.–0.028 3.7
0.013 0.013 0.013 n.d. 0
0.014 0.013 0.023 n.d.–0.034 9.1
0.013 0.013 0.013 n.d. 0
0.520 0.013 0.520 n.d.–10.2 5.0
0.014 0.013 0.015 n.d.–0.055 5.1
BPB
15.4
0.100 0.025 0.745 n.d.–0.817 10.0
0.050 0.025 0.025 n.d.–0.691 3.7
0.384 0.060 1.97 n.d.–2.37 52.9
0.123 0.025 0.471 n.d.–0.530 27.3
1.74 0.025 8.23 n.d.–11.8 45.5
0.429 0.025 2.10 n.d.–4.45 25.0
0.131 0.025 0.752 n.d.–1.34 12.8
BPF
3.27
0.013 0.013 0.013 n.d. 0
0.317 0.013 1.37 n.d.–5.67 11.1
0.037 0.013 0.095 n.d.–0.423 5.9
0.335 0.013 1.79 n.d.–3.56 9.1
0.013 0.013 0.013 n.d. 0
0.520 0.013 4.05 n.d.–5.09 15.0
0.133 0.013 0.088 n.d.–3.94 5.1
BPP
0.644
0.011 0.005 0.040 n.d.–0.073 15.0
0.054 0.005 0.104 n.d.–1.13 14.8
0.012 0.005 0.036 n.d.–0.110 11.8
0.005 0.005 0.005 n.d. 0
0.564 0.145 2.53 n.d.–4.25 72.7
2.16 0.005 2.60 n.d.–42.3 30.0
0.042 0.005 0.078 n.d.–1.11 23.1
BPS
Table 1. Concentrations (ng g–1 fresh weight) of bisphenol analogues in different categories of food items collected from nine cities in China.
Downloaded by [Ryerson University] at 00:51 03 February 2015
0.037
0.252 0.025 1.82 n.d.–2.57 15.0
0.077 0.025 0.025 n.d.–1.43 3.7
0.025 0.025 0.025 n.d. 0
0.025 0.025 0.025 n.d. 0
0.025 0.025 0.025 n.d. 0
0.066 0.025 0.373 n.d.–0.502 10.0
0.025 0.025 0.025 n.d. 0
BPZ
(continued )
27.0
8.16 1.00 14.7 0.108–134 100
7.03 0.365 46.7 n.d.–66.7 70.4
2.01 1.29 4.88 n.d.–10.9 94.1
2.69 1.25 7.95 n.d.–9.31 90.9
16.5 7.85 41.5 0.570–42.2 100
4.39 0.668 17.1 n.d.–44.6 85.0
5.97 0.507 19.9 n.d.–130 76.9
ΣBPs
322 C. Liao and K. Kannan
BPA 0.005 28.6 n.d.–127 19.0 0.005 0.005 0.005 n.d. 0 0.005 0.005 0.005 n.d. 0 0.005 0.005 0.005 n.d. 0 0.065 0.005 0.131 n.d.–1.79 6.3 0.005 0.005 0.005 n.d. 0 0.711 0.005 0.232 n.d.–127 8.3
0.005 0.005 0.005 n.d.–0.017 3.8 0.005 0.005 0.005 n.d. 0 0.017 0.005 0.069 n.d.–0.133 9.1 0.019 0.005 0.030 n.d.–0.506 14.6 0.009 0.005 0.025 n.d.–0.044 15.4 0.016 0.005 0.026 n.d.–0.759 13.1
BPAP
0.005 0.112 n.d.–0.505 21.4
BPAF
0.158 0.013 0.013 n.d.–31.5 2.4
0.013 0.013 0.013 n.d. 0
0.669 0.013 0.013 n.d.–31.5 2.1
0.047 0.013 0.201 n.d.–0.390 9.1
0.013 0.013 0.013 n.d. 0
0.013 0.013 0.013 n.d. 0
0.013 0.013 n.d. 0
BPB
2.50 0.025 1.71 n.d.–623 19.4
0.372 0.025 1.62 n.d.–2.00 23.1
0.450 0.025 3.50 n.d.–7.97 10.4
0.194 0.025 0.686 n.d.–0.709 45.5
0.025 0.025 0.025 n.d. 0
0.140 0.025 1.07 n.d.–1.61 7.7
0.025 6.01 n.d.–623 26.2
BPF
Note: n.d., Not detectable; values below the LOQ were substituted with LOQ/2 for the calculation of the mean and median.
Median 0.224 95th percentile 7.09 Range n.d.–59.6 Frequency (%) 59.5 Cookies/snacks (n = 26) Mean 4.16 Median 0.668 95th percentile 5.40 Range n.d.–78.1 Frequency (%) 73.1 Beverages (n = 4) Mean 15.5 Median 7.84 95th percentile 41.3 Range n.d.–46.4 Frequency (%) 75.0 Cooking oils (n = 11) Mean 1.92 Median 0.724 95th percentile 8.26 Range n.d.–8.87 Frequency (%) 54.5 Condiments (n = 48) Mean 7.19 Median 0.005 95th percentile 12.6 Range n.d.–299 Frequency (%) 25.0 Others (n = 13) Mean 0.445 Median 0.097 95th percentile 1.81 Range n.d.–3.02 Frequency (%) 53.8 All (n = 289) Mean 4.94 Median 0.212 95th percentile 17.4 Range n.d.–299 Frequency (%) 60.9
Table 1. Continued .
0.686 0.013 1.32 n.d.–73.1 6.9
0.067 0.013 0.295 n.d.–0.718 7.7
0.595 0.013 0.013 n.d.–28.0 2.1
0.013 0.013 0.013 n.d. 0
0.013 0.013 0.013 n.d. 0
0.091 0.013 0.013 n.d.–2.06 3.8
0.013 3.80 n.d.–73.1 16.7
BPP
Downloaded by [Ryerson University] at 00:51 03 February 2015
0.287 0.005 0.241 n.d.–42.3 22.5
0.009 0.005 0.027 n.d.–0.034 15.4
0.020 0.005 0.115 n.d.–0.245 16.7
0.014 0.005 0.056 n.d.–0.108 9.1
0.005 0.005 0.005 n.d. 0
0.065 0.005 0.302 n.d.–0.624 34.6
0.005 0.270 n.d.–24.8 31.0
BPS
0.050 0.025 0.025 n.d.–2.57 3.1
0.025 0.025 0.025 n.d. 0
0.026 0.025 0.025 n.d.–0.088 2.1
0.025 0.025 0.025 n.d. 0
0.025 0.025 0.025 n.d. 0
0.025 0.025 0.025 n.d. 0
0.025 0.025 n.d.–0.401 4.8
BPZ
9.35 0.786 43.6 n.d.–671 77.5
0.944 0.323 2.70 n.d.–3.11 69.2
9.03 0.095 26.2 n.d.–300 45.8
2.24 0.913 8.35 n.d.–8.96 72.7
15.6 7.93 41.3 n.d.–46.5 75.0
4.51 1.08 5.66 n.d.–78.2 80.8
2.02 73.0 n.d.–671 90.5
ΣBPs
Food Additives & Contaminants: Part A 323
Downloaded by [Ryerson University] at 00:51 03 February 2015
C. Liao and K. Kannan
category were canned products. The categories of “others” (mean = 0.445, 95th percentile = 1.81 ng g–1), meat and meat products (0.579, 2.26 ng g–1), milk and milk products (1.47, 3.76 ng g–1), and cooking oils (1.92, 8.26 ng g–1) contained low concentrations of BPA (Table 1). The highest concentration of BPA (299 ng g–1) was found in a spiced salt sample (processed sample) in the condiments category (ready-to-serve foods). Notable concentrations of BPA were found in baked bread pieces (130 ng g–1) and canned black rice gruel (53.7 ng g–1) in the cereals and cereal products category, crispy bean flakes (66.6 ng g–1) and canned green bean soup (55.2 ng g–1) in bean products category, hawthorn (132 ng g–1) in the fruits category, and wild seaweed biscuits (78.1 ng g–1) in the cookies/snacks category. BPF also was detected in various food items (19.4% detection rate). The categories of vegetables and fish/seafood contained high concentrations of BPF, and the overall mean concentrations in these two categories were 15.4 and 1.74 ng g–1, respectively (Table 1). The highest BPF concentration of 623 ng g–1 was found in canned braised bamboo shoots (in the vegetables category). The elevated concentration of BPF found in this sample increased the overall mean value for the vegetables category (2.50 versus 0.348 ng g–1 after the exclusion of this sample). BPS was more frequently detected in food samples than was BPF (22.5% versus 19.4% for the entire sample set), although the concentrations of BPS (mean = 0.287, range = < LOQ–42.3 ng g–1) were generally lower than the concentrations of BPF. Other bisphenol analogues were detected less frequently at concentrations below 1 ng g–1, with the exception of BPAP and BPP in vegetables, for which the respective mean concentration and detection rate were 4.68 ng g–1 and 19.0% for BPAP and 3.27 ng g–1 and 16.7% for BPP (Table 1). The concentrations of total bisphenols in foodstuffs analysed in this study (∑BPs; sum of eight bisphenols) ranged from < LOQ to 671 ng g–1 (mean = 9.35, median = 0.786 ng g–1) (Table 1). The categories of vegetables, fish and seafood, and beverages contained high ∑BP concentrations, and the respective mean and median values were 27.0 and 2.02, 16.5 and 7.85, and 15.6 and 7.93 ng g–1 (Table 1). Although the ∑BP concentrations varied widely within each food category and among the 13 categories, no significant difference in ∑BP concentrations was found between and within food categories (p > 0.05). Food samples analysed in the present study were collected from nine cities throughout China and the distribution of ∑BPs was examined for the nine cities (Figure 1; see also Table S5 in the Supplementary material). The food samples from Harbin (located in north-eastern China) contained higher ∑BP concentrations (mean = 26.6, median = 1.77 ng g–1), which were comparable with ∑BPs found in food samples from Liuzhou (southern China) (18.5 and 3.10 ng g–1; p > 0.05) but approximately one to two orders of magnitude higher than in foods from Baoding
1000 ∑ BPs concentration (ng g–1)
324
100
10
1
0.1
ng
di
o Ba
ng
iji
Be
in
rb
Ha
an
Jin
ng
ha
c Jin
ou
zh
u Li
h
ng
Qi
ao
gd
n ua
S
ai
gh
n ha
jin
an Ti
Figure 1. Comparison of total bisphenol concentrations (∑BPs) in foodstuffs from nine sampling locations in China. The lower and upper stars denote the 1st and 99th percentiles, the lower and upper whiskers represent the 5th and 95th percentiles, and the bottom and top edges of the box show the 25th and 75th percentiles, respectively. The open square and the line within the box represent the mean and median concentrations, respectively. Dots are outliers.
(0.983, 0.536 ng g–1), Beijing (1.00, 0.100 ng g–1), and Qinghuangdao (3.20, 0.724 ng g–1); nevertheless, the differences were not significant (p > 0.05). The contribution of individual bisphenol analogue (expressed as a percentage of the total) to total concentrations (∑BPs) was calculated for each category of food samples (Figure 2). Although BPA substitutes have been gradually developed (Sueiro et al. 2003; Satoh et al. 2004; Kuruto-Niwa et al. 2005; Feng et al. 2012), our previous studies suggested that BPA is still the major bisphenol found in consumer products (Liao, Liu, Alomirah, et al. 2012; Liao, Liu, Guo, et al. 2012; Liao, Liu, Moon, et al. 2012; Liao & Kannan 2013). In this study, BPA was the predominant compound in the food samples (with a 60.9% detection rate). The contribution of BPA to ∑BP concentrations in foods ranged from 43% in the meat and meat products category to 100% in the beverages category, with an overall mean value of 64% (Figure 2). BPF, BPS and BPP accounted for 10% ± 24% (mean ± SD), 7.7% ± 23%, and 6.5% ± 22% of the ∑BP concentrations. The contributions of other bisphenols to ∑BP concentrations in foods were below 5%. The overall composition profiles of bisphenol analogues in foodstuffs from China were consistent with the profiles reported for consumer products (Liao, Liu, Alomirah, et al. 2012; Liao, Liu, Guo, et al. 2012; Liao, Liu, Moon, et al. 2012; Liao & Kannan 2013). It should be noted that the predominance of censored data (concentration below the LOQ) can skew the distribution of bisphenols found at low concentrations and detection rates in foods (including BPAP, BPB and BPZ)
Food Additives & Contaminants: Part A BPA
BPAF
BPAP
BPB
BPF
BPP
BPS
325
BPZ
Cereals and cereal products Meat and meat products Fish and seafood Eggs Milk and milk products Bean products Fruits Vegetables Cookies/snacks Beverages Cooking oils Condiments
All 0%
Figure 2. China.
20%
40%
60%
80%
100%
Relative composition of individual bisphenols to total bisphenol concentrations in foodstuffs collected from nine cities in
(Table 1); for this calculation the concentrations below the LOQ were substituted with a value of zero. BPA analogues are widely used in the lining of metal cans and in the production of polycarbonate plastics (Geens et al. 2012; Huang et al. 2012; Vandenberg et al. 2012). Leaching of bisphenols from food packaging materials is considered a major source of these compounds in foods (Schecter et al. 2010). A large number of studies have reported on the leaching of BPA from packaging materials, including metallic cans and plastics, into foods (MunguiaLopez & Soto-Valdez 2001; Munguia-Lopez et al. 2002; Cao & Corriveau 2008; Cao et al. 2009; Lim et al. 2009; Viñas et al. 2010; Cooper et al. 2011; Simoneau et al. 2011; Oca et al. 2013). The food samples analysed in this study were divided into four groups (can, glass, paper and plastic), based on the type of packaging material. The canned foods contained higher mean concentrations of BPA (20.2 ng g–1), which were two to four times higher than those packaged in plastic (4.11 ng g–1) and paper (11.6 ng g–1) (p > 0.05), and two orders of magnitude higher than those packaged in glass (0.304 ng g–1; p < 0.001). A similar pattern was found for the ∑BP concentrations among the four groups (Figure 3; see also Table S6 in the Supplementary material). The non-parametric Spearman correlation analysis was used to test the correlations among the concentrations of individual bisphenols. BPAP, BPB, BPP and BPZ were detected infrequently in the food samples and, therefore, these four bisphenol analogues were not included in the dataset. Significant positive correlations were found between BPA and BPF (r = 0.23, p < 0.01), BPA and BPS (r = 0.17, p < 0.01), and BPAF and BPS (r = 0.14,
p < 0.05), suggesting co-occurrence and similarity in sources in foods (Table 2).
Comparison of BPA concentrations in foodstuffs with those reported earlier from China Very few studies have reported the concentrations of BPA in foodstuffs from China (Shao, Han, Li, et al. 2007; Shao, Han, Tu, et al. 2007; Niu et al. 2012). The detection rates of BPA in different food items analysed in our study were generally
BPA
1000 Concentration (ng g–1)
Downloaded by [Ryerson University] at 00:51 03 February 2015
Others
∑BPs
100 10 1 0.1 0.01 Canned
Glass
Paper
Plastic
Figure 3. Concentrations of BPA and ∑BPs in food samples sold in different packaging materials in China. The lower and upper stars denote the 1st and 99th percentiles, the lower and upper whiskers represent the 5th and 95th percentiles, and the bottom and top edges of the box show the 25th and 75th percentiles, respectively. The open square and the line within the box represent the mean and median concentrations, respectively. Dots are outliers.
326
C. Liao and K. Kannan Table 2.
BPA BPAF BPF BPS ΣBPs
Spearman correlations among concentrations of bisphenol analogues in foodstuffs from China. BPA
BPAF
BPF
BPS
ΣBPs
1.000 0.028 0.232** 0.169** 0.777**
1.000 0.041 0.144* 0.098
1.000 –0.008 0.430**
1.000 0.254**
1.000
Note: * and ** Correlations significant at the 0.05 and 0.01 levels (two-tailed).
Table 3. Reported concentrations of BPA in foodstuffs from China with those measured in this study.
Downloaded by [Ryerson University] at 00:51 03 February 2015
Location Beijing Beijing Beijing Beijing Beijing Beijing Guangzhou Guangzhou Hong Kong Several citiesb Several cities Several cities Several cities Several cities Several cities Several cities
Food item
Year
Beverages including bottled water Meat Eggs Milk Oils Cereals Bottled water Tap water Freshwater and marine fish Beverages Meat and meat products Eggs Milk and milk products Cooking oils Cereals and cereal products Fish and seafood
2004 2006 2006 2006 2011 2012 2009 2009 2010 2012 2012 2012 2012 2012 2012 2012
Range (ng g–1 or ng ml–1) < 0.00004a
Food Additives & Contaminants: Part A Publication details, including instructions for authors and subscription information: http://www.tandfonline.com/loi/tfac20
A survey of bisphenol A and other bisphenol analogues in foodstuffs from nine cities in China a
Chunyang Liao & Kurunthachalam Kannan
a
a
Wadsworth Center, New York State Department of Health, and Department of Environmental Health Sciences, School of Public Health, State University of New York at Albany, Empire State Plaza, Albany, NY 12201-0509, USA Accepted author version posted online: 21 Nov 2013.Published online: 30 Jan 2014.
Click for updates To cite this article: Chunyang Liao & Kurunthachalam Kannan (2014) A survey of bisphenol A and other bisphenol analogues in foodstuffs from nine cities in China, Food Additives & Contaminants: Part A, 31:2, 319-329, DOI: 10.1080/19440049.2013.868611 To link to this article: http://dx.doi.org/10.1080/19440049.2013.868611
PLEASE SCROLL DOWN FOR ARTICLE Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) contained in the publications on our platform. However, Taylor & Francis, our agents, and our licensors make no representations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of the Content. Any opinions and views expressed in this publication are the opinions and views of the authors, and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon and should be independently verified with primary sources of information. Taylor and Francis shall not be liable for any losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoever or howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use of the Content. This article may be used for research, teaching, and private study purposes. Any substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any form to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http:// www.tandfonline.com/page/terms-and-conditions
Food Additives & Contaminants: Part A, 2014 Vol. 31, No. 2, 319–329, http://dx.doi.org/10.1080/19440049.2013.868611
A survey of bisphenol A and other bisphenol analogues in foodstuffs from nine cities in China Chunyang Liao and Kurunthachalam Kannan* Wadsworth Center, New York State Department of Health, and Department of Environmental Health Sciences, School of Public Health, State University of New York at Albany, Empire State Plaza, Albany, NY 12201-0509, USA
Downloaded by [Ryerson University] at 00:51 03 February 2015
(Received 29 October 2013; accepted 19 November 2013) Bisphenol A (BPA) is a high-production-volume chemical that is widely used in polycarbonate plastics and epoxy food-can coatings. Following several studies that have reported adverse effects of BPA over the past decade, other bisphenol analogues, such as bisphenol F (BPF), bisphenol S (BPS), bisphenol AF (BPAF), and bisphenol B (BPB), have been gradually developed as substitutes for BPA in several applications. Nevertheless, few studies have reported on the occurrence of compounds other than BPA in foodstuffs. In this study, 289 food samples (13 categories: cereals and cereal products, meat and meat products, fish and seafood, eggs, milk and milk products, bean products, fruits, vegetables, cookies/snacks, beverages, cooking oils, condiments, and others), collected from nine cities in China, were analysed for eight bisphenol analogues using high-performance liquid chromatography-tandem mass spectrometry (HPLC-MS/MS). BPA and BPF were found widely in foodstuffs at concentrations ranging from below the limit of quantitation (LOQ) to 299 ng g–1 (mean = 4.94 ng g–1) and from below the LOQ to 623 ng g–1 (mean = 2.50 ng g–1), fresh weight, respectively. The highest total concentrations of bisphenols (∑BPs: sum of eight bisphenols) were found in the category of vegetables that included canned products (mean = 27.0 ng g–1), followed by fish and seafood (16.5 ng g–1) and beverages (15.6 ng g–1). ∑BP concentrations (mean = 2–3 ng g–1) in milk and milk products, cooking oils, and eggs were low. Food samples sold in metallic cans contained higher mean ∑BP concentrations (56.9 ng g–1) in comparison with those packaged in glass (0.43 ng g–1), paper (11.9 ng g–1), or plastic (6.40 ng g–1). The daily dietary intakes of bisphenols were estimated, based on the mean concentrations measured and daily consumption rates of foods, to be 646 and 664 ng kg–1 bw day–1 for men and women, respectively. Keywords: bisphenols; BPA; food; dietary exposure; market basket survey; daily intake
Introduction Bisphenol A (BPA) is widely used as a monomer in the manufacturing of polycarbonate plastics and epoxy resins, which are used in a wide range of consumer products, including plastic food containers and epoxy food-can coatings (Geens et al. 2012; Huang et al. 2012; Vandenberg et al. 2012). BPA is also used in printed circuit boards, building materials, compact discs, medical devices, dental fillings and thermal receipt papers (Geens et al. 2012; Gallart-Ayala et al. 2013). It has been estimated that approximately 3% of the polycarbonate and 10% of the epoxy resins are used in materials intended to come in contact with foods (Plastics Europe 2007). Other bisphenol analogues, including bisphenol F (BPF), bisphenol S (BPS), bisphenol AF (BPAF), and bisphenol B (BPB), have been gradually developed as substitutes for BPA in the production of polycarbonate plastics and epoxy resins (Sueiro et al. 2003; Satoh et al. 2004; Kuruto-Niwa et al. 2005; Feng et al. 2012). In comparison with BPA, BPF possesses lower viscosity and better resistance against solvents and is increasingly used in epoxy resin products (Danzl et al. 2009). BPS is widely used as a replacement for BPA in thermal receipt papers (Liao Liu, Kannan, et al. 2012). BPAF is used as a cross-linking reagent in the *Corresponding author. Email: [email protected] © 2014 Taylor & Francis
production of fluoropolymers and fluoroelastomers (Feng et al. 2012). Considerable controversy surrounds exposure to and toxicity of BPA, and this compound has been identified as an endocrine disruptor due to its potential to elicit developmental and reproductive toxicity (Geens et al. 2012; Huang et al. 2012; Vandenberg et al. 2012). Adverse effects of BPA at environmentally relevant exposure doses have been reported in laboratory animals (Vandenberg et al. 2012). Neonatal exposure to BPA of rats at 10 µg kg–1 body weight (bw) resulted in early and persistent over-expression of genes involved in DNA methylation/demethylation (Tang et al. 2012). In an in vitro study using murine cells, exposure to BPA at concentrations ranging from 10−10 to 10−8 M (approximately 0.023 to 2.3 ng ml–1) markedly induced spermatogonial proliferation (Sheng & Zhu 2011). Acute toxicity, genotoxicity and estrogenic activity of BPS, BPB and BPF have been reported (Chen et al. 2002; Rivas et al. 2002; Okuda et al. 2011; Pisapia et al. 2012). One study showed that exposure to BPAF considerably reduced testosterone levels in adult male rats (Feng et al. 2012). A few other studies have shown that BPF and BPS are more resistant to degradation in the environment than is BPA (Ike et al. 2006; Danzl et al. 2009).
Downloaded by [Ryerson University] at 00:51 03 February 2015
320
C. Liao and K. Kannan
Humans can be exposed to BPA that leaches from polycarbonate plastics or epoxy resin packaging materials into foodstuffs (Vandenberg et al. 2007). Earlier studies have shown that BPA can leach from polycarbonate drinking and baby bottles or from metallic-can epoxy coatings into various food products (Cao & Corriveau 2008; Cao et al. 2009; Cooper et al. 2011; Simoneau et al. 2011). Heat processing or acidic conditions can enhance the leaching of BPA from packaging materials or storage containers (Munguia-Lopez & Soto-Valdez 2001; Munguia-Lopez et al. 2002; Lim et al. 2009). In view of the migration of BPA from packaging materials, the European Commission has established a migration limit of 0.6 mg kg–1 for BPA in plastic materials and articles intended to come in contact with foodstuffs (European Commission 2004). Although very few studies have reported on the migration of bisphenols other than BPA from plastics and epoxy coatings (Viñas et al. 2010; Oca et al. 2013), it is expected to be similar to those reported for BPA. A few pilot studies have reported the occurrence of BPA in food items collected in China (Shao, Han, Li, et al. 2007; Shao, Han, Tu, et al. 2007; Niu et al. 2012). BPA was found in 13 of 27 (48%) meat samples, including pork, beef, lamb, chicken, duck and fish, collected from supermarkets in Beijing, China, at concentrations ranging from 0.33 to 7.08 ng g–1 wet weight (ww) (Shao, Han, Li, et al. 2007). Low concentrations and detection rates for BPA were reported in milk (10%) and eggs (30%) from China, at concentrations ranging from 0.35 to 10.5 ng g–1 ww (Shao, Han, Tu, et al. 2007). A recent study reported BPA in several cereals and cereal products from Beijing and found BPA at measurable levels (range = 1.0– 3.8 ng g–1 ww; detection rate = 21%) (Niu et al. 2012). The earlier studies were sporadic and analysed very few samples collected from localised areas in China. To our knowledge, studies on the occurrence of other bisphenol analogues in foodstuffs from China are not available. The aim of this study was to assess comprehensively dietary exposure to bisphenols by the general population in China. A total of 289 food samples were collected from nine cities in China in 2012 for the analysis of eight bisphenol analogues, including BPA, BPAF, BPB, BPF and BPS. Concentrations of bisphenols in several food categories and in foodstuffs sold in various types of packaging materials were determined and regional distribution in concentrations elucidated. On the basis of the measured concentrations in food samples and reported daily consumption rates of individual food items, the dietary intakes of bisphenols were calculated.
(BPAP; 1571-75-1; 99%), bisphenol F (BPF; 620-92-8; 98%), bisphenol P (BPP; 2167-51-3; 99%), bisphenol S (BPS; 80-09-1; 98%), and bisphenol Z (BPZ; 843-55-0; 98%) were purchased from Sigma-Aldrich (St. Louis, MO, USA). Bisphenol B (BPB; 77-40-7; 98%) was purchased from TCI America (Portland, OR, USA); and 13 C12-labelled BPA (263261-65-0; 99%) was from Cambridge Isotope Laboratories (Andover, MA, USA). Formic acid (98.2%) was purchased from Sigma-Aldrich. All organic solvents (HPLC grade), including methanol, acetonitrile, acetone, hexane and dichloromethane, were purchased from Mallinckrodt Baker (Phillipsburg, NJ, USA). Milli-Q water was prepared by an ultrapure water system (Barnstead International, Dubuque, IA, USA). Sample collection From July to September 2012, a total of 289 food samples were collected from nine cities in China, including Baoding (n = 28), Beijing (31), Harbin (41), Jinan (41), Jinchang (36), Liuzhou (49), Qinghuangdao (37), Shanghai (14) and Tianjin (12) (Figure S1 in the Supplementary material). Food samples, purchased mainly from large retail stores and a few local grocery stores, were grouped into 13 categories as cereals and cereal products (rice and rice products/wheat flour/bread/noodles/pasta/corn products), meat and meat products (beef/ pork/chicken/duck/sausages), fish and seafood (fish/ shrimp/squid), eggs (salted eggs), milk and milk products (milk/infant formula/yogurt/cheese), bean products (red and green beans/soybeans/orchid beans/dried bean curd), fruits (walnuts/chestnuts/jujubes/plums/hawthorn/raisins), vegetables (mushrooms/peanuts/peppers/seaweed, bamboo shoots/potatoes/cabbage/salted mustard), cookies/ snacks (candy/chocolate/biscuits/potato chips/pancakes), beverages (juice/liquor/coffee drinks), cooking oils (bean, peanut, corn, and sesame oils), condiments (soy sauce/ vinegar/cooking wine/ketchup/bean paste/aniseed/chilli powder), and others (jelly/black sesame powder/lotus root starch/milk tea powder/coffee powder). Various brands of foods, including national brands, store brands and specialty brands, were chosen to represent items commonly consumed by the Chinese population. Further details of the food items analysed in this study have been reported elsewhere (Liao et al. 2013). All samples were stored at –20°C until processing. Sample preparation
Materials and methods Chemicals Bisphenol A (BPA; CAS No.: 80-05-7; purity: 97%), bisphenol AF (BPAF; 1478-61-1; 97%), bisphenol AP
The extraction and clean-up procedures in the analysis of food samples have been described in detail elsewhere (Liao & Kannan 2013). For the purpose of extraction, food samples were divided into four broad categories: solid foods, cooking oils, beverages, and milk and its products. Solid foods were homogenised (using a stainless
Downloaded by [Ryerson University] at 00:51 03 February 2015
Food Additives & Contaminants: Part A steel Sorvall® Omni-Mixer; Dupont, Newtown, CT, USA), freeze dried, weighed (1.0–3.0 g dry weight), and extracted twice with acetonitrile (6 ml each). After evaporation to near dryness, the extract was redissolved in 2 ml of 10% dichloromethane/hexane and purified with a Strata® NH2 cartridge (200 mg/3 cc; Phenomenex, Torrance, CA, USA). The final elute was concentrated to 0.5 ml. The cooking oil samples (1.5–4.0 g) were extracted using the same procedure described above, without the freeze-drying step. Beverage samples were weighed (approximately 5 g) and extracted twice with ethyl acetate (6 ml each) by liquid–liquid extraction (LLE). The extract was evaporated to near dryness, redissolved in 2 ml of 10% dichloromethane/hexane (v/v), and purified with Strata® NH2 cartridge, as described above. Milk and milk products (approximately 3 g) were extracted with about 6 ml of acetonitrile (sample: acetonitrile = 1:2, v/v) (Yan et al. 2009) by LLE. After centrifugation, the supernatant was evaporated to approximately 4 ml, diluted to 10 ml with 0.2% formic acid (pH 2.5), and purified with an Oasis® MCX cartridge (60 mg/3 cc; Waters, Milford, MA, USA) (Liao & Kannan 2013). The final elute was concentrated to 0.5 ml. HPLC-MS/MS analysis A Shimadzu Prominence Series liquid chromatograph, LC-20AD system (Shimadzu USA, Canby, OR, USA) was used for the separation of target analytes. The analytical column was Betasil C18 (2.1 × 100 mm, 5 μm; Thermo Electron Corporation, Waltham, MA, USA), which was connected to a Javelin guard column (Betasil C18, 2.1 × 20 mm, 5 μm; Thermo Electron Corporation). The mobile phase was methanol and water; the elution gradient is shown in Table S1 in the Supplementary material. The mobile phase flow rate was 300 µl min–1, and 10 µl of the sample extract was injected. An Applied Biosystems API 3200 electrospray triple quadrupole mass spectrometer (ESI-MS/MS; Applied Biosystems, Foster City, CA, USA) was used for the determination of bisphenols. The MS/MS was operated in an electrospray-negative ionisation multiple reaction monitoring (MRM) mode, and the MRM transitions were optimised for each analyte (Table S2 in the Supplementary material). The MS/MS parameters are shown in Table S3 of the Supplementary material. Quality assurance and quality control (QA/QC) Identification of target analytes was based on the retention times relative to 13C12-BPA. Quantification was based on the linear regression curve (r > 0.99) generated from a 12point calibration standard at concentrations in the range of 0.01–100 ng ml–1. The measured concentrations were corrected for the recoveries of the internal standard. For
321
every 60 food samples analysed, procedural blanks (n = 3), spiked blanks (n = 3), and spiked matrixes (n = 4) were analysed to evaluate possible contamination and matrix effects from sample preparation steps and instrumental analysis. Procedural blanks contained trace levels of BPA (0.048 ng g–1), BPF (0.067 ng g–1), and BPS (0.005 ng g–1), and these values were subtracted (from mean values found in procedural blanks for each batch of analysis) from sample values. The recoveries of standards spiked in blanks (20 ng for each analyte; mean ± SD) ranged from 62% ± 7% (BPS) to 107% ± 18% (BPB). The average recoveries of standards spiked in food matrixes (20 ng for each analyte), from four broad categories of foods, were in the range of 79–109% for solid foods, 62–120% for cooking oils, 78–122% for beverages, and 66–126% for milk and milk products (Table S4 in the Supplementary material). The limit of quantitations (LOQs), calculated from the value of the lowest acceptable calibration standard and a nominal sample weight of 1.0 g, were 0.01 ng g–1 ww for BPA, BPAF, BPAP and BPS, 0.025 ng g–1 ww for BPB and BPP, and 0.05 ng g–1 ww for BPF and BPZ (Table S2 in the Supplementary material). A calibration standard and methanol were injected after every 10 samples on the instrument to check for instrumental background, carryover and stability.
Data analysis Liquid samples were weighed prior to extraction. The moisture contents of solid food samples were determined for the conversion of data from a dry-weight to a wetweight basis. All data are presented on a wet-weight (or fresh-weight) basis, unless specified otherwise. For statistical analysis, values below the LOQ were set to half the LOQ (LOQ/2). Spearman’s correlation analysis and oneway analysis of variance (ANOVA) with the Tukey test were conducted using SPSS 17.0 and Origin 7.5. A value of p < 0.05 was considered as being statistically significant.
Results and discussion Bisphenol concentrations BPA was frequently (detection rate = 60.9%) found in food samples from China at concentrations ranging from < LOQ to 299 ng g–1 fresh weight (mean = 4.94 ng g–1) (Table 1). The highest BPA concentrations (mean = 15.5, 95th percentile = 41.3 ng g–1) were found in beverages; this was followed, in decreasing order, by fish and seafood (14.1, 40.2 ng g–1), fruits (7.76, 14.5 ng g–1), and condiments (7.19, 12.6 ng g–1). Canned foods contributed a relatively high mean concentration of BPA in these food categories; in particular, three of four samples in the beverages
Cereals and cereal products (n = 39) Mean 5.60 Median 0.383 95th percentile 18.6 Range n.d.–130 Frequency (%) 69.2 Meat and meat products (n = 20) Mean 0.579 Median 0.098 95th percentile 2.26 Range n.d.–3.14 Frequency (%) 55.0 Fish and seafood (n = 11) Mean 14.1 Median 4.46 95th percentile 40.2 Range 0.327–42.1 Frequency (%) 100 Eggs (n = 11) Mean 2.15 Median 0.967 95th percentile 7.53 Range n.d.–9.22 Frequency (%) 63.6 Milk and milk products (n = 17) Mean 1.47 Median 0.942 95th percentile 3.76 Range n.d.–10.8 Frequency (%) 82.4 Bean products (n = 27) Mean 6.48 Median 0.147 95th percentile 46.1 Range n.d.–66.6 Frequency (%) 63.0 Fruits (n = 20) Mean 7.76 Median 0.530 95th percentile 14.5 Range n.d.–132 Frequency (%) 85.0 Vegetables (n = 42) Mean 2.88
BPA 0.027 0.005 0.033 n.d.–0.575 5.1 0.102 0.005 0.343 n.d.–1.67 10.0 0.006 0.005 0.011 n.d.–0.017 9.1 0.010 0.005 0.030 n.d.–0.047 18.2 0.063 0.005 0.352 n.d.–0.746 11.8 0.029 0.005 0.035 n.d.–0.622 7.4 0.007 0.005 0.019 n.d.–0.031 10.0 4.68
0.012 0.005 0.054 n.d.–0.097 10.0 0.088 0.005 0.446 n.d.–0.759 36.4 0.020 0.005 0.075 n.d.–0.117 36.4 0.005 0.005 0.005 n.d. 0 0.006 0.005 0.009 n.d.–0.014 7.4 0.008 0.005 0.018 n.d.–0.023 25.0 0.031
BPAP
0.005 0.005 0.005 n.d.–0.016 2.6
BPAF
0.013
0.013 0.013 0.013 n.d. 0
0.013 0.013 0.013 n.d.–0.028 3.7
0.013 0.013 0.013 n.d. 0
0.014 0.013 0.023 n.d.–0.034 9.1
0.013 0.013 0.013 n.d. 0
0.520 0.013 0.520 n.d.–10.2 5.0
0.014 0.013 0.015 n.d.–0.055 5.1
BPB
15.4
0.100 0.025 0.745 n.d.–0.817 10.0
0.050 0.025 0.025 n.d.–0.691 3.7
0.384 0.060 1.97 n.d.–2.37 52.9
0.123 0.025 0.471 n.d.–0.530 27.3
1.74 0.025 8.23 n.d.–11.8 45.5
0.429 0.025 2.10 n.d.–4.45 25.0
0.131 0.025 0.752 n.d.–1.34 12.8
BPF
3.27
0.013 0.013 0.013 n.d. 0
0.317 0.013 1.37 n.d.–5.67 11.1
0.037 0.013 0.095 n.d.–0.423 5.9
0.335 0.013 1.79 n.d.–3.56 9.1
0.013 0.013 0.013 n.d. 0
0.520 0.013 4.05 n.d.–5.09 15.0
0.133 0.013 0.088 n.d.–3.94 5.1
BPP
0.644
0.011 0.005 0.040 n.d.–0.073 15.0
0.054 0.005 0.104 n.d.–1.13 14.8
0.012 0.005 0.036 n.d.–0.110 11.8
0.005 0.005 0.005 n.d. 0
0.564 0.145 2.53 n.d.–4.25 72.7
2.16 0.005 2.60 n.d.–42.3 30.0
0.042 0.005 0.078 n.d.–1.11 23.1
BPS
Table 1. Concentrations (ng g–1 fresh weight) of bisphenol analogues in different categories of food items collected from nine cities in China.
Downloaded by [Ryerson University] at 00:51 03 February 2015
0.037
0.252 0.025 1.82 n.d.–2.57 15.0
0.077 0.025 0.025 n.d.–1.43 3.7
0.025 0.025 0.025 n.d. 0
0.025 0.025 0.025 n.d. 0
0.025 0.025 0.025 n.d. 0
0.066 0.025 0.373 n.d.–0.502 10.0
0.025 0.025 0.025 n.d. 0
BPZ
(continued )
27.0
8.16 1.00 14.7 0.108–134 100
7.03 0.365 46.7 n.d.–66.7 70.4
2.01 1.29 4.88 n.d.–10.9 94.1
2.69 1.25 7.95 n.d.–9.31 90.9
16.5 7.85 41.5 0.570–42.2 100
4.39 0.668 17.1 n.d.–44.6 85.0
5.97 0.507 19.9 n.d.–130 76.9
ΣBPs
322 C. Liao and K. Kannan
BPA 0.005 28.6 n.d.–127 19.0 0.005 0.005 0.005 n.d. 0 0.005 0.005 0.005 n.d. 0 0.005 0.005 0.005 n.d. 0 0.065 0.005 0.131 n.d.–1.79 6.3 0.005 0.005 0.005 n.d. 0 0.711 0.005 0.232 n.d.–127 8.3
0.005 0.005 0.005 n.d.–0.017 3.8 0.005 0.005 0.005 n.d. 0 0.017 0.005 0.069 n.d.–0.133 9.1 0.019 0.005 0.030 n.d.–0.506 14.6 0.009 0.005 0.025 n.d.–0.044 15.4 0.016 0.005 0.026 n.d.–0.759 13.1
BPAP
0.005 0.112 n.d.–0.505 21.4
BPAF
0.158 0.013 0.013 n.d.–31.5 2.4
0.013 0.013 0.013 n.d. 0
0.669 0.013 0.013 n.d.–31.5 2.1
0.047 0.013 0.201 n.d.–0.390 9.1
0.013 0.013 0.013 n.d. 0
0.013 0.013 0.013 n.d. 0
0.013 0.013 n.d. 0
BPB
2.50 0.025 1.71 n.d.–623 19.4
0.372 0.025 1.62 n.d.–2.00 23.1
0.450 0.025 3.50 n.d.–7.97 10.4
0.194 0.025 0.686 n.d.–0.709 45.5
0.025 0.025 0.025 n.d. 0
0.140 0.025 1.07 n.d.–1.61 7.7
0.025 6.01 n.d.–623 26.2
BPF
Note: n.d., Not detectable; values below the LOQ were substituted with LOQ/2 for the calculation of the mean and median.
Median 0.224 95th percentile 7.09 Range n.d.–59.6 Frequency (%) 59.5 Cookies/snacks (n = 26) Mean 4.16 Median 0.668 95th percentile 5.40 Range n.d.–78.1 Frequency (%) 73.1 Beverages (n = 4) Mean 15.5 Median 7.84 95th percentile 41.3 Range n.d.–46.4 Frequency (%) 75.0 Cooking oils (n = 11) Mean 1.92 Median 0.724 95th percentile 8.26 Range n.d.–8.87 Frequency (%) 54.5 Condiments (n = 48) Mean 7.19 Median 0.005 95th percentile 12.6 Range n.d.–299 Frequency (%) 25.0 Others (n = 13) Mean 0.445 Median 0.097 95th percentile 1.81 Range n.d.–3.02 Frequency (%) 53.8 All (n = 289) Mean 4.94 Median 0.212 95th percentile 17.4 Range n.d.–299 Frequency (%) 60.9
Table 1. Continued .
0.686 0.013 1.32 n.d.–73.1 6.9
0.067 0.013 0.295 n.d.–0.718 7.7
0.595 0.013 0.013 n.d.–28.0 2.1
0.013 0.013 0.013 n.d. 0
0.013 0.013 0.013 n.d. 0
0.091 0.013 0.013 n.d.–2.06 3.8
0.013 3.80 n.d.–73.1 16.7
BPP
Downloaded by [Ryerson University] at 00:51 03 February 2015
0.287 0.005 0.241 n.d.–42.3 22.5
0.009 0.005 0.027 n.d.–0.034 15.4
0.020 0.005 0.115 n.d.–0.245 16.7
0.014 0.005 0.056 n.d.–0.108 9.1
0.005 0.005 0.005 n.d. 0
0.065 0.005 0.302 n.d.–0.624 34.6
0.005 0.270 n.d.–24.8 31.0
BPS
0.050 0.025 0.025 n.d.–2.57 3.1
0.025 0.025 0.025 n.d. 0
0.026 0.025 0.025 n.d.–0.088 2.1
0.025 0.025 0.025 n.d. 0
0.025 0.025 0.025 n.d. 0
0.025 0.025 0.025 n.d. 0
0.025 0.025 n.d.–0.401 4.8
BPZ
9.35 0.786 43.6 n.d.–671 77.5
0.944 0.323 2.70 n.d.–3.11 69.2
9.03 0.095 26.2 n.d.–300 45.8
2.24 0.913 8.35 n.d.–8.96 72.7
15.6 7.93 41.3 n.d.–46.5 75.0
4.51 1.08 5.66 n.d.–78.2 80.8
2.02 73.0 n.d.–671 90.5
ΣBPs
Food Additives & Contaminants: Part A 323
Downloaded by [Ryerson University] at 00:51 03 February 2015
C. Liao and K. Kannan
category were canned products. The categories of “others” (mean = 0.445, 95th percentile = 1.81 ng g–1), meat and meat products (0.579, 2.26 ng g–1), milk and milk products (1.47, 3.76 ng g–1), and cooking oils (1.92, 8.26 ng g–1) contained low concentrations of BPA (Table 1). The highest concentration of BPA (299 ng g–1) was found in a spiced salt sample (processed sample) in the condiments category (ready-to-serve foods). Notable concentrations of BPA were found in baked bread pieces (130 ng g–1) and canned black rice gruel (53.7 ng g–1) in the cereals and cereal products category, crispy bean flakes (66.6 ng g–1) and canned green bean soup (55.2 ng g–1) in bean products category, hawthorn (132 ng g–1) in the fruits category, and wild seaweed biscuits (78.1 ng g–1) in the cookies/snacks category. BPF also was detected in various food items (19.4% detection rate). The categories of vegetables and fish/seafood contained high concentrations of BPF, and the overall mean concentrations in these two categories were 15.4 and 1.74 ng g–1, respectively (Table 1). The highest BPF concentration of 623 ng g–1 was found in canned braised bamboo shoots (in the vegetables category). The elevated concentration of BPF found in this sample increased the overall mean value for the vegetables category (2.50 versus 0.348 ng g–1 after the exclusion of this sample). BPS was more frequently detected in food samples than was BPF (22.5% versus 19.4% for the entire sample set), although the concentrations of BPS (mean = 0.287, range = < LOQ–42.3 ng g–1) were generally lower than the concentrations of BPF. Other bisphenol analogues were detected less frequently at concentrations below 1 ng g–1, with the exception of BPAP and BPP in vegetables, for which the respective mean concentration and detection rate were 4.68 ng g–1 and 19.0% for BPAP and 3.27 ng g–1 and 16.7% for BPP (Table 1). The concentrations of total bisphenols in foodstuffs analysed in this study (∑BPs; sum of eight bisphenols) ranged from < LOQ to 671 ng g–1 (mean = 9.35, median = 0.786 ng g–1) (Table 1). The categories of vegetables, fish and seafood, and beverages contained high ∑BP concentrations, and the respective mean and median values were 27.0 and 2.02, 16.5 and 7.85, and 15.6 and 7.93 ng g–1 (Table 1). Although the ∑BP concentrations varied widely within each food category and among the 13 categories, no significant difference in ∑BP concentrations was found between and within food categories (p > 0.05). Food samples analysed in the present study were collected from nine cities throughout China and the distribution of ∑BPs was examined for the nine cities (Figure 1; see also Table S5 in the Supplementary material). The food samples from Harbin (located in north-eastern China) contained higher ∑BP concentrations (mean = 26.6, median = 1.77 ng g–1), which were comparable with ∑BPs found in food samples from Liuzhou (southern China) (18.5 and 3.10 ng g–1; p > 0.05) but approximately one to two orders of magnitude higher than in foods from Baoding
1000 ∑ BPs concentration (ng g–1)
324
100
10
1
0.1
ng
di
o Ba
ng
iji
Be
in
rb
Ha
an
Jin
ng
ha
c Jin
ou
zh
u Li
h
ng
Qi
ao
gd
n ua
S
ai
gh
n ha
jin
an Ti
Figure 1. Comparison of total bisphenol concentrations (∑BPs) in foodstuffs from nine sampling locations in China. The lower and upper stars denote the 1st and 99th percentiles, the lower and upper whiskers represent the 5th and 95th percentiles, and the bottom and top edges of the box show the 25th and 75th percentiles, respectively. The open square and the line within the box represent the mean and median concentrations, respectively. Dots are outliers.
(0.983, 0.536 ng g–1), Beijing (1.00, 0.100 ng g–1), and Qinghuangdao (3.20, 0.724 ng g–1); nevertheless, the differences were not significant (p > 0.05). The contribution of individual bisphenol analogue (expressed as a percentage of the total) to total concentrations (∑BPs) was calculated for each category of food samples (Figure 2). Although BPA substitutes have been gradually developed (Sueiro et al. 2003; Satoh et al. 2004; Kuruto-Niwa et al. 2005; Feng et al. 2012), our previous studies suggested that BPA is still the major bisphenol found in consumer products (Liao, Liu, Alomirah, et al. 2012; Liao, Liu, Guo, et al. 2012; Liao, Liu, Moon, et al. 2012; Liao & Kannan 2013). In this study, BPA was the predominant compound in the food samples (with a 60.9% detection rate). The contribution of BPA to ∑BP concentrations in foods ranged from 43% in the meat and meat products category to 100% in the beverages category, with an overall mean value of 64% (Figure 2). BPF, BPS and BPP accounted for 10% ± 24% (mean ± SD), 7.7% ± 23%, and 6.5% ± 22% of the ∑BP concentrations. The contributions of other bisphenols to ∑BP concentrations in foods were below 5%. The overall composition profiles of bisphenol analogues in foodstuffs from China were consistent with the profiles reported for consumer products (Liao, Liu, Alomirah, et al. 2012; Liao, Liu, Guo, et al. 2012; Liao, Liu, Moon, et al. 2012; Liao & Kannan 2013). It should be noted that the predominance of censored data (concentration below the LOQ) can skew the distribution of bisphenols found at low concentrations and detection rates in foods (including BPAP, BPB and BPZ)
Food Additives & Contaminants: Part A BPA
BPAF
BPAP
BPB
BPF
BPP
BPS
325
BPZ
Cereals and cereal products Meat and meat products Fish and seafood Eggs Milk and milk products Bean products Fruits Vegetables Cookies/snacks Beverages Cooking oils Condiments
All 0%
Figure 2. China.
20%
40%
60%
80%
100%
Relative composition of individual bisphenols to total bisphenol concentrations in foodstuffs collected from nine cities in
(Table 1); for this calculation the concentrations below the LOQ were substituted with a value of zero. BPA analogues are widely used in the lining of metal cans and in the production of polycarbonate plastics (Geens et al. 2012; Huang et al. 2012; Vandenberg et al. 2012). Leaching of bisphenols from food packaging materials is considered a major source of these compounds in foods (Schecter et al. 2010). A large number of studies have reported on the leaching of BPA from packaging materials, including metallic cans and plastics, into foods (MunguiaLopez & Soto-Valdez 2001; Munguia-Lopez et al. 2002; Cao & Corriveau 2008; Cao et al. 2009; Lim et al. 2009; Viñas et al. 2010; Cooper et al. 2011; Simoneau et al. 2011; Oca et al. 2013). The food samples analysed in this study were divided into four groups (can, glass, paper and plastic), based on the type of packaging material. The canned foods contained higher mean concentrations of BPA (20.2 ng g–1), which were two to four times higher than those packaged in plastic (4.11 ng g–1) and paper (11.6 ng g–1) (p > 0.05), and two orders of magnitude higher than those packaged in glass (0.304 ng g–1; p < 0.001). A similar pattern was found for the ∑BP concentrations among the four groups (Figure 3; see also Table S6 in the Supplementary material). The non-parametric Spearman correlation analysis was used to test the correlations among the concentrations of individual bisphenols. BPAP, BPB, BPP and BPZ were detected infrequently in the food samples and, therefore, these four bisphenol analogues were not included in the dataset. Significant positive correlations were found between BPA and BPF (r = 0.23, p < 0.01), BPA and BPS (r = 0.17, p < 0.01), and BPAF and BPS (r = 0.14,
p < 0.05), suggesting co-occurrence and similarity in sources in foods (Table 2).
Comparison of BPA concentrations in foodstuffs with those reported earlier from China Very few studies have reported the concentrations of BPA in foodstuffs from China (Shao, Han, Li, et al. 2007; Shao, Han, Tu, et al. 2007; Niu et al. 2012). The detection rates of BPA in different food items analysed in our study were generally
BPA
1000 Concentration (ng g–1)
Downloaded by [Ryerson University] at 00:51 03 February 2015
Others
∑BPs
100 10 1 0.1 0.01 Canned
Glass
Paper
Plastic
Figure 3. Concentrations of BPA and ∑BPs in food samples sold in different packaging materials in China. The lower and upper stars denote the 1st and 99th percentiles, the lower and upper whiskers represent the 5th and 95th percentiles, and the bottom and top edges of the box show the 25th and 75th percentiles, respectively. The open square and the line within the box represent the mean and median concentrations, respectively. Dots are outliers.
326
C. Liao and K. Kannan Table 2.
BPA BPAF BPF BPS ΣBPs
Spearman correlations among concentrations of bisphenol analogues in foodstuffs from China. BPA
BPAF
BPF
BPS
ΣBPs
1.000 0.028 0.232** 0.169** 0.777**
1.000 0.041 0.144* 0.098
1.000 –0.008 0.430**
1.000 0.254**
1.000
Note: * and ** Correlations significant at the 0.05 and 0.01 levels (two-tailed).
Table 3. Reported concentrations of BPA in foodstuffs from China with those measured in this study.
Downloaded by [Ryerson University] at 00:51 03 February 2015
Location Beijing Beijing Beijing Beijing Beijing Beijing Guangzhou Guangzhou Hong Kong Several citiesb Several cities Several cities Several cities Several cities Several cities Several cities
Food item
Year
Beverages including bottled water Meat Eggs Milk Oils Cereals Bottled water Tap water Freshwater and marine fish Beverages Meat and meat products Eggs Milk and milk products Cooking oils Cereals and cereal products Fish and seafood
2004 2006 2006 2006 2011 2012 2009 2009 2010 2012 2012 2012 2012 2012 2012 2012
Range (ng g–1 or ng ml–1) < 0.00004a
